cdtMerging: CDT merging methods
In rijaf-iri/CDTMergingF: Climate Data Tools Merging Methods

Description Usage Arguments Examples

View source: R/cdtMerging.R

This function blends station data with a gridded data. The blending methods are "Regression Kriging", "Simple Bias Adjustment" (additive bias), "Cressman Scheme" and "Barnes Scheme". The radius of influence for the interpolation method varies for each grid point.

cdtMerging(time.step = "dekadal", start.date = c(1981, 1, 1),
  end.date = c(2018, 12, 3), station = list(file = NULL, sep = ",",
  miss = "-99"), netcdf = list(dir = NULL, varid = NA, order.lon = 1,
  order.lat = 2, format = "rfe_%s%s%s.nc"), merging.method = "SBA",
  interp.method = "idw", spheric = FALSE, maxdist = 2.5,
  pass.ratio = c(1, 0.5, 0.2), pass.nmin = c(5, 4, 3),
  pass.nmax = c(15, 10, 7), neg.value = FALSE, output = list(dir =
  NULL, format = "rr_mrg_%s%s%s.nc", name = "precip", units = "mm",
  longname = NA, prec = "short"), use.RnoR = FALSE,
  pars.RnoR = list(wet.day = 1, smooth = FALSE), vgm.model = c("Exp",
  "Gau", "Sph", "Pen"), parallel = list(dopar = TRUE, detect.cores =
  TRUE, nb.cores = 2))

`time.step`	Time step of the data. Should be "daily", "pentad", "dekadal" or "monthly".
`start.date`	A vector of the start date to merge. Daily data: `c(year, month, day)` Pentad data: `c(year, month, pentad)`, `pentad` must be 1, 2, 3, 4, 5, or 6 Dekadal data: `c(year, month, dekad)`, `dekad` must be 1, 2 or 3 Monthly data: `c(year, month)`
`end.date`	A vector of the start date to merge. Same format as `start.date`.
`station`	A list containing the station information in the form `list(file, sep, miss)`. `file`: full path to the station file (CDT station data format) `sep`: separator of the column `miss`: missing value flag
`netcdf`	A list containing the netcdf data information in the form `list(dir, order.lon, order.lat, format)`. `dir`: full path to the directory containing the ne netcdf data `varid`: name of the variable to be used `order.lon`: order of the longitude (netcdf dimension order) `order.lat`: order of the latitude `format`: filename format of the netcdf files
`merging.method`	Merging method to be used. Valid options: "CSc", "BSc", "SBA" or "RK". "CSc": Cressman Scheme "BSc": Barnes Scheme "SBA": Simple Bias Adjustment "RK": Regression Kriging
`interp.method`	Interpolation method to be used for "SBA" and "RK". Valid options: "idw", "shepard", "spheremap" or "kriging". "idw": Inverse distance weighted "shepard": Modified Shepard interpolation "spheremap": Spheremap interpolation method "kriging": Ordiranry kriging
`spheric`	If `FALSE` (default), then a cartesian distance will be computed. If set to `TRUE`, a spherical distance (using a standard great circle method) will be computed.
`maxdist`	Maximum radius of influence in decimal degree.
`pass.ratio`	A vector giving the fraction of `maxdist` to be used for each pass.
`pass.nmin`	A vector giving the minimum number of stations to be used to interpolate a grid point for each pass. Must be the same length as `pass.ratio`.
`pass.nmax`	A vector giving the maximum number of stations to be used to interpolate a grid point for each pass. Must be the same length as `pass.ratio`.
`neg.value`	If `TRUE`, negative values will be kept. If `FALSE` (default), negative values will be set to zero.
`output`	A list of the output information. `dir`: full path to the directory to save the results `format`: filename format of the output file `name`: name of the netcdf variable. Example: "precip" or "temp" `units`: units of the netcdf variable `longname`: longname of the netcdf `prec`: Precision of the output created variable. Valid options: "short", "integer", "float" or "double".
`use.RnoR`	If `TRUE`, apply rain-no-rain mask for rainfall data.
`pars.RnoR`	A list of parameters to be used if `use.RnoR` is `TRUE`. `wet.day`: wet day definition `smooth`: if `TRUE`, smooth the mask
`vgm.model`	A vector of variogram model to be used if `interp.method` is "kriging". Default is `c("Exp", "Gau", "Sph", "Pen")`.
`parallel`	A list of parameters for parallel processing. `dopar`: if `TRUE` (default), use parallel computing `detect.cores`: if `TRUE` (default), detect the number of CPU cores. `nb.cores`: if `dopar` is `TRUE` and `detect.cores` is `FALSE`, the number of CPU cores to be used

## Not run: 
cdtMerging(
     time.step = "dekadal",
     start.date = c(2018, 1, 1),
     end.date = c(2018, 12, 3),
     station = list(file = "~/DONNEES/STN_DATA/CDT_RR_dekad_1981-2018_CLM.csv",
                 sep = ",", miss = "-99"),
     netcdf = list(dir = "~/DONNEES/GRIDDED_DATA/tamsat_v3/dekadal", varid = "rfe",
                 order.lon = 1, order.lat = 2, format = "rfe_%s%s%s.nc"),
     merging.method = "SBA", 
     interp.method = "idw",
     spheric = TRUE,
     maxdist = 1.0,
     pass.ratio = c(1, 0.75, 0.5),
     pass.nmin = c(5, 4, 3),
     pass.nmax = c(15, 10, 7),
     neg.value = FALSE,
     output = list(dir = "~/DONNEES/OUT_Merging", format = "rr_mrg_%s%s%s.nc",
                 name = "precip", units = "mm",
                 longname = "Merged Station-Satellite Rainfall", prec = "short"),
     use.RnoR = TRUE,
     pars.RnoR = list(wet.day = 1.0, smooth = FALSE),
     vgm.model = c("Exp", "Gau", "Sph", "Pen")
 )

## End(Not run)